SELFEXPLAIN: A Self-Explaining Architecture for Neural Text Classifiers

This paper proposes a globally interpretable layer and a locally interpretable layer to identify the most influential concepts in the whole training set and in a given sample respectively. They use non-terminals as candidates and use their representations (averaged through training set or no-average) to optimize task loss.


  • Good to see constituency trees are still helpful at least for interpretation.
  • Design is good.
  • Their way to build a representation of the input without contribution of \mathbf{u}_j is not sounding:g(\mathbf{u}_j)-g(\mathbf{u}_s) is still tied to \mathbf{u}_j. Why don’t you try some masking?
  • 5: Transformative: This paper is likely to change our field. It should be considered for a best paper award.
  • 4.5: Exciting: It changed my thinking on this topic. I would fight for it to be accepted.
  • 4: Strong: I learned a lot from it. I would like to see it accepted.
  • 3.5: Leaning positive: It can be accepted more or less in its current form. However, the work it describes is not particularly exciting and/or inspiring, so it will not be a big loss if people don’t see it in this conference.
  • 3: Ambivalent: It has merits (e.g., it reports state-of-the-art results, the idea is nice), but there are key weaknesses (e.g., I didn’t learn much from it, evaluation is not convincing, it describes incremental work). I believe it can significantly benefit from another round of revision, but I won’t object to accepting it if my co-reviewers are willing to champion it.
  • 2.5: Leaning negative: I am leaning towards rejection, but I can be persuaded if my co-reviewers think otherwise.
  • 2: Mediocre: I would rather not see it in the conference.
  • 1.5: Weak: I am pretty confident that it should be rejected.
  • 1: Poor: I would fight to have it rejected.

0 投票人

1 Like